search

GoogleCloudPlatform / training-data-analyst

Labs and demos for courses for GCP Training (http://cloud.google.com/training).
Apache License 2.0
7.86k stars 5.86k forks source link

Unable to complete "AK8S-11 Configuring Pod Autoscaling and NodePools" #1288

Open exofaaste opened 3 years ago

exofaaste commented 3 years ago

Currently trying to complete the quicklabs "AK8S-11 Configuring Pod Autoscaling and NodePools" instruction is:

  1. git clone https://github.com/GoogleCloudPlatform/training-data-analyst.git
  2. cd ~/training-data-analyst/courses/ak8s/v1.1/11_Autoscaling/
  3. kubectl create -f web.yaml --save-config
  4. kubectl expose deployment web --target-port=8080 --type=NodePort

As per the commit the new commands would be as simple as to change point # 2 to this

  1. cd ~/training-data-analyst/courses/ak8s/v1.1/Autoscaling/

Lab is unable to check progress and is marking 0% completed. Error Message "It doesn't look like you've completed this step yet. Try again."

abursavich commented 3 years ago

I left this comment on the lab when I finished it, but I'm adding it here too:

This lab is broken and very frustrating.

The deployment must specify CPU resource requests/limits in order for the autoscaler to function. Resources are not specified in the given YAML.

I have no idea how I finally got part three to succeed... I think it was because I ramped up the load and scaled the webs up to 4 replicas on the preemptible nodes, but that wasn't part of the step's directions.